0. Software Defined Infrastructure
18 April 2017
Saqib Ahmad Khan
Country General Manager, GBM Pakistan

1. Organizations Are Embracing New Opportunities
Cloud
New consumption models
80% of new applications will include cloud delivery or deployment
Big Data, Analytics (HPC)
Ability to gain insights quickly
2.5 billion GB of data are being generated every day
Social & Mobile
Explosive growth of
file and object data
500 million Tweets a day; 95% of mobile traffic is data

2. Foundation for Optimizing New IT Environments
Social & Mobile
Big Data & Analytics
Other Business Apps
Optimization of performance and agility of the application lifecycle
Cloud Environment
Traditional
Environment
Automation via open standards and policy driven patterns
Abstraction of capabilities and service delivery from physical infrastructure
Software Defined Infrastructure
(Server, Storage, Network)

3. Software Defined Infrastructure
Accelerate results and lower costs with a scalable, efficient environment
Save money, time and resources
Transform a static IT infrastructure into a dynamic resource, workload and data-aware environment
Securely integrate mobile services
Use data efficiently for analytics services
Manage in an always-on world for service predictability

4. Software Defined Infrastructure
Accelerate results and lower costs with a scalable, efficient environment
Save money, time and resources
Transform a static IT infrastructure into a dynamic resource, workload and data-aware environment
Securely integrate mobile services
Use data efficiently for analytics services
Manage in an always-on world for service predictability
Application workloads are serviced automatically by the most appropriate resource running locally, in the cloud, or in a hybrid cloud environment

5. SDI for Scale Out Applications
Example
Applications
High Performance Analytics (Low Latency Parallel)
Homegrown
Hadoop / Big Data
High Performance Computing (Batch, Serial, MPI, Workflow)
Application Frameworks (Long Running Services)
Do you have any Scale Out application?
Do you have resource sharing to address utilization?
Are you looking to improve time-to-results?

6. Scheduling & Acceleration With Infrastructure Sharing
Example
Applications
High Performance Analytics (Low Latency Parallel)
Homegrown
Hadoop / Big Data
High Performance Computing (Batch, Serial, MPI, Workflow)
Application Frameworks (Long Running Services)
IBM Spectrum Symphony
MapReduce
(Symphony)
IBM Spectrum LSF
IBM Spectrum Conductor
Workload
Engines
IBM Spectrum Computing
Resource
Management

7. Heterogeneous Infrastructure Support
High Performance Analytics (Low Latency Parallel)
Hadoop / Big Data
High Performance Computing (Batch, Serial, MPI, Workflow)
Application Frameworks (Long Running Services)
IBM Spectrum Symphony
MapReduce
(Symphony)
IBM Spectrum LSF
IBM Spectrum Conductor
Workload
Engines
IBM Spectrum Computing
Resource
Management
Hypervisor
x86
Linux on z
On-premise, On-cloud, Hybrid Infrastructure
(heterogeneous distributed computing and storage environment)

8. Infrastructure Models
Bare Metal, Private Cloud, Public / Hybrid Cloud
Bare Metal Provisioning
SoftLayer APIs & Services
Virtual Machine Provisioning
Infrastructure
Management
IBM Spectrum Cluster Foundation
Cloud
IBM Cloud Manager with OpenStack
Hypervisor
x86
Linux on z
On-premise, On-cloud, Hybrid Infrastructure
(heterogeneous distributed computing and storage environment)

9. Data Management Combined with Compute
File, Object, Block
Data
Management
Elastic Storage, SAN Volume Controller
Bare Metal Provisioning
SoftLayer APIs & Services
Virtual Machine Provisioning
Infrastructure
Management
IBM Spectrum Cluster Foundation
Cloud
IBM Cloud Manager with OpenStack
Hypervisor
x86
Linux on z
On-premise, On-cloud, Hybrid Infrastructure
(heterogeneous distributed computing and storage environment)

10. Breadth and Depth of Solution
High Performance Analytics (Low Latency Parallel)
Hadoop / Big Data
High Performance Computing (Batch, Serial, MPI, Workflow)
Application Frameworks (Long Running Services)
Commercial
Applications
IBM Spectrum Computing
MapReduce
(Symphony)
IBM Spectrum Symphony
IBM Spectrum Conductor
IBM Spectrum LSF
Other Management Software
Workload
Engines
Resource
Management
Data
Management
Elastic Storage, SAN Volume Controller
Bare Metal Provisioning
SoftLayer APIs & Services
Virtual Machine Provisioning
Infrastructure
Management
IBM Spectrum Cluster Foundation
Cloud
IBM Cloud Manager with OpenStack
Hypervisor
x86
Linux on z
On-premise, On-cloud, Hybrid Infrastructure
(heterogeneous distributed computing and storage environment)

11. Software Defined Storage
Spectrum
Control
Analytics-driven data management to reduce management costs by up to 50%
Spectrum Protect
Optimized data protection to reduce backup costs by up to 38%
Spectrum Archive
Fast data retention that reduces TCO for active archive data by up to 90%
Spectrum Virtualize
Virtualization of mixed environments stores up to 5x more data
Accelerate
Enterprise storage for cloud deployed in minutes instead of months
Scale
High-performance highly scalable storage for unstructured data

12. Software Defined Storage Benefits
Agility
Increased flexibility through support of heterogeneous hardware
Open standards enable software control across multiple platforms
Cost
Efficiency
Match workload requirements to the appropriate storage
Provides massive, virtually limitless scalability
Control and QoS
Automated, policy-driven management helps drive lower cost and operational efficiencies

13. Balancing Organizational Needs
IT managers: reduce costs while maintaining service levels
End users: gain competitive advantage
CIO & CFO: increase business agility while managing costs

14. Next Steps?
What are the organization priorities?
Business Perspective – What is the potential return on investment?
Technology Perspective – What is the current infrastructure and future roadmap?
Can we create a plan in a discovery workshop?

15. Thank You